Multimodal & Grounded benchmark
ExtractBench (LlamaIndex) leaderboard
ExtractBench: schema-guided enterprise document extraction (LlamaIndex leaderboard). Every model the catalog carries a published ExtractBench (LlamaIndex) value for, ranked by that value.
CategoryMultimodal & Grounded
MeasureOrder-insensitive value F1 overall and by document length; word and page grounding F1; cost and latency
Tasks370 documents, 4,869 pages, 8 business domains, 67 document types, each with a JSON Schema
DifficultyShort documents are near solved; long record lists and grounding separate systems
Scores schema-guided extraction from 370 enterprise documents and 4,869 pages on value F1, record completeness, word and page grounding and cost, split by short, medium and long documents.
About ExtractBench (LlamaIndex)
What the benchmark measures and how its value is published, from the source’s own record.
What is ExtractBench (LlamaIndex)?
ExtractBench (LlamaIndex) (ExtractBench: schema-guided enterprise document extraction (LlamaIndex leaderboard)) is a multimodal & grounded benchmark published by ExtractBench: A Benchmark for Schema-Guided Enterprise Document Extraction. Scores schema-guided extraction from 370 enterprise documents and 4,869 pages on value F1, record completeness, word and page grounding and cost, split by short, medium and long documents. Its evaluation set is described by the source as 370 documents, 4,869 pages, 8 business domains, 67 document types, each with a JSON Schema. The source rates its difficulty as short documents are near solved; long record lists and grounding separate systems.
How is ExtractBench (LlamaIndex) scored?
Each model's result is published by ExtractBench: A Benchmark for Schema-Guided Enterprise Document Extraction as order-insensitive value F1 overall and by document length; word and page grounding F1; cost and latency. ModelScale prints that value to 2 decimal places, exactly as published, and never converts or rescales it. This page ranks the 24 models that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.
ModelScale
